Uncovering Local Spatial Context via Multiscale Geographically Weighted Binomial Regression
This paper introduces a novel Multiscale Geographically Weighted Binomial Regression (MGWBR) model, extending the existing Multiscale Geographically Weighted Regression (MGWR) framework to accommodate binomial response variables. By integrating a Local Scoring Algorithm (LSA), the proposed MGWBR model accounts for spatial non-stationarity in binary outcomes and enables the estimation of covariate-specific bandwidths and localized parameter estimates. This approach addresses key limitations of traditional global binomial models and standard GWR models, which typically apply a single bandwidth and overlook the potential for spatially varying relationships across different predictors. To evaluate the performance of MGWBR, the model is applied to two simulated datasets and two empirical datasets: one on landslide occurrences in Clearwater national forest park and one on voting behavior in the Southeast U.S. during the 2020 presidential election. Results from both the simulated and real-world applications demonstrate that MGWBR provides improved model fit over traditional global binary regression models and effectively captures spatial heterogeneity in covariate effects, highlighting its potential as a robust tool for spatial analysis involving binary data.
Authors
- Alexander Stewart Fotheringham (ORCID: https://orcid.org/0000-0002-0407-1901)
- Mehak Sachdeva (ORCID: https://orcid.org/0000-0003-4375-5422)
- Chen-Lun Kao (ORCID: https://orcid.org/0009-0002-6998-790X)
Publication Details
- Journal
- Figshare
- Published
- 2026-11-21
- DOI
- https://doi.org/10.6084/m9.figshare.30142678
- Primary Topic
- Spatial and Panel Data Analysis
- Type
- article
- Field-Weighted Citation Impact
- 0.00